Data Scientist II, ML Infrastructure
Build and productionize ML measurement, causal inference, and platform tooling at Pinterest. Translate research into scalable pipelines, develop self-serve causal tools, and create centralized systems for feature importance, model evaluation, and infrastructure efficiency.
About the job
What you’ll do
- Translate research-grade DS workflows (e.g., proxy metrics, staleness models) into production ML pipelines using Airflow, WandB & Ray while establishing reusable patterns for other teams.
- Apply and productionize causal inference methods using the production ML stack (propensity scoring, IPW, TMLE) to address high-stakes measurement questions beyond experimental capabilities. Build self-serve tooling to empower non-experts to derive rigorous causal insights at scale.
- Partner with ML engineers and product teams to identify opportunities for improved tooling, metrics, and measurement methods, unlocking step-change improvements in model quality and business outcomes.
- Leverage Pinterest's rich metadata and engagement signals to build data-driven frameworks, from feature importance to content deindexing, that improve platform efficiency and speed.
- Design and build centralized ML platform tooling to improve feature and model creation, evaluation, and trust, including production systems that operate daily at scale across all models.
What we’re looking for
- 2+ years of hands-on experience as an applied scientist, ML engineer, research scientist or software engineer, with significant ML production experience.
- Strong Python skills; experience with PyTorch or equivalent deep learning frameworks; familiarity with distributed compute (Spark, Ray). Ray specifically is a strong plus.
- Enthusiasm for building tools and platforms that multiply the impact of an entire ML organization; not just solving one-off problems.
- Deep ML theory knowledge with extremely strong fundamentals that can help us reason about ML models from first principles.
- Proficiency in software development best practices including version control, code review, and reproducible ML pipelines.
- Experience with workflow management tools (Airflow, Prefect, Jenkins, or similar) for reliable ML pipeline orchestration.
- Bachelor’s/Master’s degree in a relevant field such as Computer Science, or equivalent experience.
Skills
Python, PyTorch, Airflow, Ray, Spark, Causal Inference, Ml Pipelines, Wandb
Similar jobs
ML Engineering jobs
Simple AI
San Francisco, CA
Build and deploy production voice AI agents for customers, creating demos, debugging edge cases, improving performance, and translating feedback into product improvements. The role combines hands-on engineering, customer engagement, and pre- and post-sales delivery.
PathAI
Boston, MADevelop and deploy machine learning models for biomedical research and AI product development, collaborating with scientific, engineering, and product teams. Requires a master's degree with 2–4 years of experience or a PhD with 0–2 years, plus strong Python and ML development skills.
DataVisor
Mountain View, CA
Builds high-scale data pipelines, distributed systems, and AI agent workflows using LLMs for fraud intelligence platform. Requires 2+ years software engineering, Python proficiency, big data tools, AWS/K8s, and ML foundations.
Skydio
San Mateo, CA
Research and develop computer vision and deep learning algorithms for autonomous drones, taking ownership of projects from prototyping through product integration. The internship requires strong C++ or Python and PyTorch skills, mathematical foundations, and software engineering ability.
Benchling
San Francisco, CA
Build datasets, evaluations, and scalable data systems that improve frontier AI models on challenging biological and scientific tasks. The role partners with scientists and AI labs and requires at least two years of experience applying biology and AI, plus hands-on LLM experience.